Haploid Bio-Induction in Plant through Mock Sexual Reproduction

Summary

Haploidization is invaluable for basic genetic research and crop breeding. The haploid bio-induction principle is an important topic that remains largely unexplored. In this study, both CenH3 RNAi and in vitro inhibition were used to simulate and induce haploids in allopolyploid crop. Notably, in vitro CenH3 inhibition showed that the results were much the same to that of RNAi in phenotype, chromosome behavior, microspore production, and haploid induction. Cytological analyses of RNAi and inhibitor-treated progenies revealed elimination of chromosomes, defective microspores with empty nuclei, thereby giving rise to pseudo male gametes, and haploid parthenogenesis induction. We found distinct defective empty microspores that were positively correlated with the decrease of CenH3 during RNAi manipulation. Investigation through both in vivo and in vitro studies revealed that haploidization was induced through the pseudo male gamete-mediated mock sexual reproduction. The present results provide insights for the haploid parthenogenesis induction process.

Graphical Abstract

Highlights

• •CenH3 RNAi and in vitro inhibition can similarly induce haploids
• •We provide the cytological basis underlying parthenogenesis induction and haploids production
• •We report pseudo male gamete-mediated mock sexual reproduction during haploid parthenogenesis
• •We provide insights for the haploid parthenogenesis induction process

Biological Sciences; Plant Biology; Plant Cytology

Introduction

Utilization of haploids in breeding procedures could significantly shorten the selection time or fix heterosis. Parthenogenesis and haploidy production are unique and crucial reproductive processes that naturally exist in plants. In the 1960s, Indian scholars Guha and Maheshwari (1964) cultured Daturametel L. anthers in vitro and successfully induced haploid plants. Mejza et al. (1993) successfully obtained the haploid plants of anther cultured wheat. Later, numerous optimization studies were performed to increase the haploid generation efficiency (Zhao et al., 2015). And the maize line Stock6 and derivatives can induce haploids when crossed with other strains (Coe, 1959; Lashermes and Beckert, 1988; Khakwani et al., 2015). Kasha and Kao (1970) reported high-frequency barley haploids in progenies of the cross Hordeum vulgare and H. bulbosum, which are caused by H. bulbosum centromeric loss of CENH3 protein and uniparental genome elimination, called the Bulbosum Technique (BT) (Sanei et al., 2011). Additionally, selective elimination of maize chromosomes has been reported in wheat × maize (Laurie and Bennett, 1988) and oat × maize (Marcínska et al., 2013) crosses and has been used in breeding cultivate varieties of these crops.

The study by Ravi and Chan (2010) showed that Arabidopsis thaliana haploid plants can be easily generated through seeds by manipulating a single centromere protein, the centromere-specific histone H3 (CenH3) protein, which provided the zygotic mitosis model underlying selective elimination of chromosomes and has opened up new opportunities for constructing haploid inducer (HI) systems through genetic engineering. These HI lines can produce haploids without in vitro culture, and CenH3 can be found in all eukaryotes. Therefore, this type of HI can be applied to breeding in almost all plants. The CenH3-mediated haploid induction technique has so far been demonstrated and tested in plants (Ravi and Chan, 2010; Ravi et al., 2010; Lermontova et al., 2011; Kuppu et al., 2015; Karimi-Ashtiyani et al., 2015; Kelliher et al., 2016).

Seed haploid generation by manipulating the conserved CenH3 has been demonstrated and elaborated in Arabidopsis. However, many crop species are recalcitrant to generate haploids, and the haploid bio-induction complex process and principle remain largely unexplored. The cytological basis of chromosome elimination and the haploid induction process in cotton were investigated in this study. Haploid inducers mediated by the histone GhCenH3 genes were obtained and identified through an in vitro inhibitor treatment strategy and an in vivo RNAi technique. Notably, our study revealed the cytological basis of haploid bio-induction through pseudo male gamete-mediated mock sexual reproduction, which provides a practical biotechnology guide for crop haploid breeding. The findings in this study highlight the importance of mock sexual reproduction for understanding parthenogenesis bio-induction and haploid production in crops.

Results and Discussion

Molecular Characteristics of Transgenic and Inhibitor-Treated Lines

In order to uncover the cytological basis and events underlying plant haploid bio-induction, RNAi and in vitro CenH3 inhibition were used to simulate and induce haploids in allopolyploid cotton. The selfing progenies of identified RNAi transformants will be investigated for further expression and functional studies. Among 23 different detected T2 positive transformant lines, the relative abundance of native GhCenH3 transcript in the RNAi line A4-43-20 was reduced the most (over 80%) based on quantitative real-time RT-PCR (Figure 1A). The follow-up characterization and analysis will be carried out on the line.

Figure 1.

Suppression of Native GhCenH3 Expression and Defective Microspore Production in Transgenic RNAi Lines and the Inhibitors Condition

(A) Suppression of native CenH3 genes of flower anthers by RNAi transgenics.

(B) The expression of native CenH3 genes of flower anthers under inhibitor treatment.

(C) Defective microspore production of RNAi transformants.

(D) Defective microspore production of CenH3 inhibitor-treated plants.

See also Figures S1 and S2. Data are represented as mean ± SD of three measurements.

Additionally, the previously reported four related inhibitors that reduce CenH3 expression and simulate the RNAi process were used in the study: CYC (Cycloheximide) (Lopes et al., 2011), ROS (Roscovitine) (Mȕller et al., 2014), MG115 (Z-leu-leu-norvalinal) (Lermontova et al., 2013), and MG132 (Z-Leu-Leu-Leu-CHO) (Moreno-Moreno et al., 2006). A microinjector was used to inject the inhibitor solution into cotton young flower buds during meiosis. The present study found that the four CenH3 inhibitors had efficient inhibitory effects on GhCenH3 expression in cotton. CYC had the greatest inhibitory effect with over 90% decreased expression, followed by MG132, ROS, and MG115 (Figure 1B), compared with controls. Based on previous studies, complete silencing of conserved CenH3 (Figure S1) is lethal. Therefore, severe but not complete suppressed status, such as that in the A4-43-20 line and CYC inhibition treatment plant (Figure S2), is ideal for further investigation.

Phenotype Analysis of Transgenic and Inhibitor-Treated Lines

The RNAi A4-43-20 plants were further inspected during their growth and development. The results exhibited that the growth and development of RNAi transformant were strongly inhibited (Figure S3). Additionally, stereoscopic microscope images of the flowers of GhCenH3 RNAi transformant showed significant size reduction (Figures 2A–2D), with very short filaments (Figures 2B–2D), small anthers in the flower buds, and abnormally dehiscent anthers (Figures 2E–2H), with less pollen in the anthers (Figures 2I and 2J) and partially defective pseudo pollen in the anthers (Figures 2K and 2L) compared with the wild-type plants. The CenH3 related inhibitor in vitro treatment lines showed similar phenotypes as the RNAi transformant (Figures S4 and S5).

Figure 2.

The Floral Phenotypic Traits of GhCenH3 RNAi Transformant in Cotton

(A) The RNAi line A4-43-20 corolla compared with the wild-type.

(B) The RNAi line stamens and pistils compared with the wild-type.

(C) The wild-type stamens and pistils.

(D) The transformant stamens and pistils.

(E) The wild-type dehiscent anthers.

(F–H) The partially dehiscent anthers of the RNAi transformant. (F) longitudinal view; (G) Horizontal view; (H) enlarged view of G.

(I) The anther of the RNAi transformant stained by I₂-KI solution.

(J) The wild-type anther stained by I₂-KI solution.

(K) The RNAi transformant pollen stained by I₂-KI solution.

(L) The wild-type pollen stained by I₂-KI solution. The defective pollen marked by red arrows.

See also Figures S3–S5.

Chromosome Elimination and Cytological Analysis during Male Meiosis

Interference of chromosome behavior was further investigated using both in vivo and in vitro strategies. Analysis of meiotic products during male meiosis of the GhCenH3 RNAi transformant showed that CenH3 reduction results in disturbed meiosis: an unequal amount of chromosomes in two plates of anaphase II (Figure 3B) and a high frequency of lagging chromosomes during anaphase II (Figure 3B and 3C), compared with wild-type (Figure 3A). The chromosome behaviors in the inhibitor-treated lines showed that inhibited CenH3 expression caused defective meiosis with chromosome behaviors quite similar to that of the RNAi transformant (Figure 3D).

Figure 3.

Chromosome Behaviors and Specific Defective Microspore Production after Disturbed Meiosis in CenH3 RNAi Transformant and Inhibitor Treatment Line in Cotton

(A) Anaphase II of the wild-type plant.

(B and C) Anaphase II of the CenH3 RNAi line A4-43-20. (B) Field of typical cell one; (C) Field of typical cell two.

(D) Anaphase II of inhibitor-treated plants. Lagging chromosomes were marked by red arrows.

(E) Normal microspores in wild-type.

(F and G) Defective microspores with empty nuclei in RNAi, marked by red stars. (F) Field of typical cell one.

(G) Field of typical cell two.(H) Normal microspores in wild-type mock treatment. (I and J) Defective microspores produced by inhibitor treatment. (I) Field of typical cell one; (J) Field of typical cell two. All cells have a complete callose protective layer.

Interestingly, we detected abundant specific defective microspores with empty nuclei in both CenH3 inhibitor treatment and RNAi conditions (with a similar ratio above 80%, Figures 3E–3J). And the results showed a positive correlation between the defective microspore formation and CenH3 decrease (Figure 1). The higher the reduction of CenH3, the higher number of defective microspores. Some microspores had shallower staining (Figures 3E–3J); we speculate that unequal meiosis leads to a decrease in chromosomes.

Combining the above findings, the RNAi transformant and inhibitor treatment line have similar characteristics. The CenH3 RNAi and inhibitor treatment results indicate that CenH3 reduction results in disturbed meiosis and produces defective microspores.

RNAi Haploid Progenies Induction and Identification

To determine the ploidy status and chromosome behavior of haploid progenies, mitotic preparations were made using root meristematic tissue of haploid progenies from selfing CenH3 RNAi (Figure S6). Analysis of haploid progeny mitosis in selfing RNAi showed that CenH3 reduction can also result in disturbed mitosis: a high frequency of chromosomes were stranded on the equatorial plate, which then became lagging chromosomes during mitosis anaphase (Figure 4A), compared with the normal haploid progenies generated from maternal wild-type plants induced by RNAi lines (Figure 4B). This result suggests that there is a steady chromosome status in bio-induced haploid in wild-type plants but a chromosome elimination effect due to reduced CenH3 in selfing RNAi progenies. And the haploid status of bio-induced haploid progenies from wild-type plants by chromosome counting was shown (2n = 26 chromosomes) in Figure 4C.

Figure 4.

Cytology of Haploid Progeny Plants Derived from the CenH3 RNAi Transformants

(A) The behavior of chromosomes during mitosis anaphase in haploid progenies from the RNAi line A4-43-20 selfing.

(B) The behavior of chromosomes during mitosis anaphase in haploid progenies from maternal wild-type haploid bio-induction.

(C) The chromosome counts of the haploid progenies produced from maternal wild-type haploid bio-induction.

See also Figure S6.

Lermontova et al. (2011) found that CenH3 RNAi in Arabidopsis thaliana resulted in dwarf plants, decreased fertility, and changed ploidy levels in somatic cells. Kelliher et al. (2016) also found that the CenH3 RNAi transgenic plants of maize were seriously affected in growth and development and they also produced haploids after selfing. The CenH3 RNAi interference research of Brassica juncea showed slow plant growth, dwarf plants, seed sterility phenomena, and genetic offspring can induce haploids and aneuploid plants by selfing (Watts et al., 2016). In addition, the aneuploids were detected in the A. thaliana CenH3 RNAi transformants (Lermontova et al., 2011). Considering the aneuploids, our results are consistent with those of Ravi and Chan (2010) and Watts et al. (2016) but differ from those of Kelliher et al. (2016) who found no aneuploids in CenH3-mediated HI lines of maize.

The application of haploid reproductive technology has the potential to revolutionize crop breeding. Haploids have been conventionally induced mainly through the generation of plants from cultivated gametophic cells and tissues by in vitro culture technologies or through the selective loss of a parental chromosome set by inter- or intraspecific hybridization in vivo (Kalinowska et al., 2019). However, owing to the application limits of current haploid technologies, scientists and plant breeders are highly interested in underlying principles as well as methodological improvements of haploidization. Recently, seed haploid induction technologies like CenH3 manipulation and haploid parthenogenesis induction have been developed. The haploid progenies could be obtained from the maternal wild-types, induced by RNAi (♀WT×RNAi♂). In this study, the CenH3 transgenic RNAi line was generated to suppress native CenH3 to simulate haploid inducer (HI) lines with relative high productivity of haploid progenies ~8%. However, to this date, the principle controlling haploidization and parthenogenesis remains almost completely unknown. Very little progress has yet been made so far on elucidating the mechanism of haploid formation and parthenogenesis.

A deeper knowledge of the haploid parthenogenesis process in plants is required. This knowledge is a significant prerequisite to understand the formation and control of the apomictic development process from sexual reproduction. The in vitro CenH3 inhibition pollinator used to simulate the RNAi HI can generate the haploid progenies. We note that the inhibitors have a transient effect, different from the continued chromosome elimination effect in transgenic RNAi approach during early zygotic embryogenesis. Therefore, haploid progenies from ♀WT×inhibition (inhibitor-treated pollinator)♂ could be generated only from female gametes stimulated with defective pseudo male gametes. It suggested that haploid progenies could be similarly produced from mock sexual reproduction using RNAi HI line as pollinator of wild-type parents. CenH3-meditated haploid induction has been conventionally considered through chromosome elimination that typically takes place during early zygotic embryogenesis. Besides this basic principle, our results revealed a notable parthenogenesis engineering process by mock sexual reproduction for haploid bio-induction, which is distinct from the model of genome elimination during zygotic mitosis (Ravi and Chan, 2010; Maheshwari et al., 2015). It can be inferred that those two processes could occur alternatively or simultaneously during CenH3-meditated haploid induction. This study updates the haploid bio-induction principle and provides a practical biotechnology guide for haploid production in crops. It is of great fundamental and practical importance in crop science, holding significant promise for its advancement in engineering of apomixis in crop breeding. Future research is necessary to explain how mock sexual reproduction is established and to elucidate the molecular mechanisms regulating haploid parthenogenesis. Except the present strategy in this study, selectively modifying and editing key amino acids in the CenH3 will be complementary to build HAI system as reported in A. thaliana (Karimi-Ashtiyani et al., 2015; Kuppu et al., 2015). Furthermore, the improvement and implementation of haploidization approaches in crop breeding remains a mission to be fulfilled.

Limitations of the Study

• •The difference between natural and bio-induced parthenogenesis/apomixis has not been investigated extensively so far.
• •Further research is needed to elucidate the molecular mechanisms regulating haploid parthenogenesis mediated by mock sexual reproduction.
• •More efficient haploid induction approaches for crop breeding need to be investigated in the future based on this study.

Resource Availability

Lead Contact

Further information should be directed and will be fulfilled by the Lead Contact, Fanchang Zeng (fczeng@sdau.edu.cn).

Materials Availability

Available from the Lead Contact under the public/local regulation and Material Transfer Agreements (MTAs) for education and research purpose if they are used not for commercial.

Data and Code Availability

The data can be found and is available within the manuscript and Supplemental Information.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Published: July 24, 2020